Activating mutations in the B-catenin oncogene are found in a wide range of common human cancers, including colorectal carcinoma, hepatocellular carcinoma, endometrial carcinoma, thyroid carcinoma, and others. Activated oncogenes such as B-catenin are an appealing cancer drug target, since they represent a required genetic gain-of-function found in cancer cells but not in the patient's normal tissues. Consistent with this idea, we have previously experimentally demonstrated that targeting oncogenic B-catenin is likely to be an effective anticancer strategy (Mol Can Ther 1:1355, 2002). In this application we propose to implement a cell-based screen to identify small molecule and natural product lead compounds that are specifically cytotoxic/cytostatic to human colon cancer cells harboring an allele of oncogenic B-catenin. To do this we will employ a paired set of gene-targeted human cancer cells that differ only in the presence or absence of their endogenous allele of mutant (oncogenic) B-catenin. Since the cells are otherwise isogenic, compounds that specifically kill or inhibit the growth of parental cells (harboring oncogenic B-catenin) but not the derivatives in which oncogenic B-catenin has been deleted by gene targeting are good candidates as B-catenin targeted lead compounds. Such hits will be further validated in a variety of secondary screens to demonstrate reproducibility and generalizability. Once feasibility has been demonstrated by screening approximately 4000 compounds, we will apply to transfer the screen to an NIH Roadmap screening center for high throughput screening efforts.